Issue metadata
Sign in to add a comment
|
--disable-gpu is ignored if --skip-gpu-data-loading is used
Reported by
rog...@opera.com,
Mar 7 2016
|
||||||||||||||||||||||
Issue descriptionVersion: 51.0.2671.0 (content_shell built locally) OS: Linux What steps will reproduce the problem? 1. ./content_shell --no-sandbox --disable-gpu --skip-gpu-data-loading chrome:gpu What is the expected output? What do you see instead? I expect hardware acceleration to be turned off, that is how it used to work, but now it is always enabled as long as --skip-gpu-data-loading is provided. This is not an important regression, but since it is a change in behavior I want to highlight it. Please use labels and text to provide additional information. This regressed when https://codereview.chromium.org/1547793004/ landed, more specifically due to the change made here: https://codereview.chromium.org/1547793004/diff/620001/content/browser/gpu/gpu_data_manager_impl_private.cc. That change postpones GpuDataManagerImplPrivate::DisableHardwareAcceleration() until GpuDataManagerImplPrivate::InitializeImpl() is called. The problem is that GpuDataManagerImplPrivate::Initialize() earlies out if --skip-gpu-data-loading is provided. I.e. the postponed call to DisableHardwareAcceleration() is never performed because GpuDataManagerImplPrivate is never initialized.
,
Mar 8 2016
Thx for reporting the issue. Please try the CL https://codereview.chromium.org/1773073003/ . Worth to copy past the following from content_switches.cc : --- // Skip gpu info collection, blacklist loading, and blacklist auto-update // scheduling at browser startup time. // Therefore, all GPU features are available, and about:gpu page shows empty // content. The switch is intended only for layout tests. // TODO(gab): Get rid of this switch entirely. const char kSkipGpuDataLoading[] = "skip-gpu-data-loading"; -- Especially: "Get rid of this switch entirely." Also it is said: "about:gpu page shows empty" but for me it is not a white page but maybe what is meant is no gpu info ? Which is the case with the suggested fix.
,
Mar 8 2016
I can confirm that the CL above fixes the problem.
,
Mar 9 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/67487fb9f97e4835d80fe1baa084a3cb97df7784 commit 67487fb9f97e4835d80fe1baa084a3cb97df7784 Author: j.isorce <j.isorce@samsung.com> Date: Wed Mar 09 20:41:24 2016 Make sure to run post init tasks even if kSkipGpuDataLoading is set Fixes regression introduced by https://codereview.chromium.org/1547793004/ which postpones call to GpuDataManagerImplPrivate::SetGLStrings() and GpuDataManagerImplPrivate::DisableHardwareAcceleration() until GpuDataManagerImplPrivate::Initialize() is called. A solution is to call RunPostInitTasks() for any early return in GpuDataManagerImplPrivate::Initialize(). R=zmo@chromium.org BUG= 592435 TEST=content_shell --no-sandbox --disable-gpu --skip-gpu-data-loading chrome:gpu Review URL: https://codereview.chromium.org/1773073003 Cr-Commit-Position: refs/heads/master@{#380207} [modify] https://crrev.com/67487fb9f97e4835d80fe1baa084a3cb97df7784/content/browser/gpu/gpu_data_manager_impl_private.cc [modify] https://crrev.com/67487fb9f97e4835d80fe1baa084a3cb97df7784/content/browser/gpu/gpu_data_manager_impl_private.h [modify] https://crrev.com/67487fb9f97e4835d80fe1baa084a3cb97df7784/content/test/gpu/page_sets/gpu_process_tests.py
,
Mar 16 2016
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by zmo@chromium.org
, Mar 7 2016Status: Assigned (was: Untriaged)